We are now posting detailed instructions for generating libraries appropriate for use with DISCOVAR. Instructions for generating 250 base reads on the HiSeq 2500 will be posted as soon as we have a version that we’re sure is portable.
You can now download DISCOVAR compatible, high coverage, 250 base PCR-free reads for the trio of NA12878 (daughter), NA12891 (father) and NA12892 (mother) from the 1000 Genomes Data Coordination Center. With these data you will be able to run DISCOVAR to call variants on any region of the genome for the trio. Please be sure to read the release and publication policies governing these data.
Many people have asked if they can use their existing Illumina datasets with DISCOVAR – datasets that don’t meet the recommendations of ~60x coverage by 250 base paired reads from a ~700 bp PCR-free fragment library. We investigated and made some minor changes to the algorithm, embodied in release 46382 onwards, and it is now possible to use shorter reads from PCR libraries – with some caveats. We have successfully tested DISCOVAR on 100 base reads from a ~180 bp PCR fragment library, obtaining reasonable results but inferior to those generated from the recommended data. For more information on please see our FAQ.
We are pleased to announce that DISCOVAR is now available to download .
DISCOVAR is a variant caller and genome assembler from the Broad Institute. It uses the latest low cost sequencing data, and can generate highly accurate variant calls for individual humans, or assemble small genomes de novo (with support for large genomes to follow). We expect it will be particularly valuable for understanding human Mendelian disease, but equally suited to investigating the biology of other organisms.